[Core] Dynamic image size support for VLMs (#5276)

Signed-off-by: Xiaowei Jiang <xwjiang2010@gmail.com>
Co-authored-by: Xiaowei Jiang <xwjiang2010@gmail.com>
Co-authored-by: ywang96 <ywang@roblox.com>
Co-authored-by: xwjiang2010 <87673679+xwjiang2010@users.noreply.github.com>
Co-authored-by: Roger Wang <136131678+ywang96@users.noreply.github.com>
This commit is contained in:
Cyrus Leung
2024-07-02 20:34:00 -07:00
committed by GitHub
co-authored by Xiaowei Jiang ywang96 xwjiang2010 Roger Wang
parent 482045ee77
commit 9831aec49f
38 changed files with 1453 additions and 664 deletions
+9 -9
View File
@@ -1,9 +1,4 @@
from transformers import AutoImageProcessor
from transformers.image_processing_utils import BaseImageProcessor
from vllm.logger import init_logger
logger = init_logger(__name__)
from typing import cast
def get_image_processor(
@@ -11,10 +6,15 @@ def get_image_processor(
*args,
trust_remote_code: bool = False,
**kwargs,
) -> BaseImageProcessor:
):
"""Gets an image processor for the given model name via HuggingFace."""
# don't put this import at the top level
# it will call torch.cuda.device_count()
from transformers import AutoImageProcessor
from transformers.image_processing_utils import BaseImageProcessor
try:
processor: BaseImageProcessor = AutoImageProcessor.from_pretrained(
processor = AutoImageProcessor.from_pretrained(
processor_name,
*args,
trust_remote_code=trust_remote_code,
@@ -34,4 +34,4 @@ def get_image_processor(
else:
raise e
return processor
return cast(BaseImageProcessor, processor)